Results 1 to 12 of 12
  1. #1
    ABW Ambassador affiliatemakeover's Avatar
    Join Date
    January 18th, 2005
    Location
    Cleveland, OH
    Posts
    821
    I'm not sure if this is news to anyone except me, but if you're interested in feedback from a staffer at Wordtracker about their beliefs about Overture, please read below.

    This response was given to me because I asked why a phrase on Overture's search term suggestion tool showed up as being searched for 90000 times, but on Wordtracker, it only shows up as being searched for 2000 times.

    Wordtracker claims that the bid checking services outside of Overture are over inflating their searches.

    Could this be true? Could all of these bid checkers be inflating the numbers so much? If they are, how is that considered "integrity" within the walls of Overture. They must find a way to stop that from happening. Opinions? I gotta tell you, I'm in this business, and I'm confused.

    Regards,

    Jim Kukral



    Hi,

    We get this question quite a lot so I have to provide you with the answer we give to our customers which is as follows:

    Our predictions are based on daily 24 hour figures. The count is for the last 60 days - but the prediction, the column on the right is for the current 24 hour period for all engines. So for example, with the keyword web hosting we have:

    web hosting 10133 10582

    which means 10,133 searches were made on Dogpile/Metacrawler over last 60 days - about 3% of the market. The 10582 is our estimate for no. searches performed per day for all engines. If Google gets about 23% of the market, then that is about 2400 searches per day or 73,000 per month. Or 146,000 over 60 days.

    Now let me throw some more information at you:

    The numbers at Overture have been extremely over-inflated by the number of different partners that they have.

    We recently carried out a test of Overture's numbers for a very random keyword '"micr toner cartridge" which turned up 52,597 in September.

    What I did was to contact everyone who had the top five positions for this keyword these are the responses I received:

    1. "I wish! Our number of visitors was less than 70 for the month.

    2. "Not anywhere close to that..."

    3. "About 2 hits a day on that word"

    We too have many keyword positions where the figures just don't match up at all. "Keywords" is a great example if we were to look at goto's stats we should be receiving 1271 a day. We receive about 2 per day.

    Our conclusion is not only is the suggestion tool skewed by the words on the front of Overture, ranking software checking the positions within the engine eg.webposition, harcoded queries in places such as epilot. But also all the bid optimizers have appeared recently. They consistently check rankings inside the pay per click engines. Artificially inflating the values and skewing the results. There are over 15 bid checkers now. You can imagine what these are doing to the results.

    This is why we will always use metacrawlers as the results will never be distorted in this way.

    So what we have is an estimate of about 146,000 for that two month period for web hosting for Google vs. their 1.5 million searches. I reckon that they're reporting impressions, rather than searches to make themselves look good. So if we're talking about 4 impressions per page (every call to the server) that puts us up to about 0.5 million searches. So we're getting into the ballpark.

    The important thing to realise is that we have ALL searches for Metacrawler/Dogpile for that 60 day period and we KNOW that they get about 3% of the market. So we have some pure facts to work with and make predictions about the other engines.

    Hope this clears things up

    Regards,

    Mike

    Wordtacker

    [ 06-25-2002: Message edited by: AffiliateMakeover.com ]

  2. #2
    Member
    Join Date
    January 18th, 2005
    Posts
    96
    This means you pay 10 times too much to Overture?

    Jonny

  3. #3
    ABW Ambassador
    Join Date
    January 18th, 2005
    Posts
    2,341
    I have heard others on this forum complaining about Overtures inflation of search numbers.

  4. #4
    ABW Ambassador
    Join Date
    January 18th, 2005
    Location
    Canada
    Posts
    1,650
    <BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>This means you pay 10 times too much to Overture?<HR></BLOCKQUOTE> No, it leads you to expect a lot more traffic from Overture than you might actually get.

  5. #5
    Member
    Join Date
    January 18th, 2005
    Posts
    96
    That's the same.
    Jonny :confused:

  6. #6
    ABW Ambassador
    Join Date
    January 18th, 2005
    Location
    Canada
    Posts
    1,650
    No it's not.

    Make sure you're thinking about the right numbers, Jonny. The numbers being questioned here are those in the search term suggestion tool which are supposed to tell you how many searches there were in the whole Overture system for this or that keyword last month.

    The issue here is that in some cases an unknown portion of those searches were done by someone's rank-checking program, not a human searcher. The problem is that this leads to inaccurate expectations about how much traffic one could expect from an Overture campaign.

    This is not at all the same as an individual marketer getting billed for bogus clicks, which is what I think you're thinking about.

    Elisabeth Archambault

  7. #7
    ABW Ambassador affiliatemakeover's Avatar
    Join Date
    January 18th, 2005
    Location
    Cleveland, OH
    Posts
    821
    Still though, does anyone else think this stinks?

    In theory, if a word is highly searched for according to their search tool, then it is more valuable to advertisers.

    So...if it's more valuable to advertisers, won't that mean that advertisers will bid higher to be at the top of it.

    So, in essence, a bogus count could possibly inflate advertisers costs because the higher the count, the more sought after the phrase is.

    At least that's how I see it. I've contacted Overture for an official response and I'll let you know.

  8. #8
    ABW Ambassador
    Join Date
    January 18th, 2005
    Location
    Canada
    Posts
    1,650
    It may lead to some unrealistic bidding, but advertisers who survive in PPC marketing have to learn how to bid according to what the traffic is actually worth to them, and not be seduced into a bidding war. Indeed, if you're expecting 90,000 visitors from a term instead of just 900 or something, it's more important than ever to make sure you are not bidding higher than what is cost-effective.

    The situation is frustrating, but the potential for problems would be worse if one got a lot more traffic than expected and had bid too much!

  9. #9
    Member
    Join Date
    January 18th, 2005
    Posts
    96
    Of course it is the same. I have to pay for nonrealitic traffic predictions which are 10 times higher than real. So I bid higher in expecting getting traffic which never would come. If I know my conversion ratio and take this ratio as a messure to figure out if advertising would work and the base of my decission is 10 times higher than it would be at the end, I pay 10 times too much for my bid.
    Jonny [img]tongue.gif[/img] [img]tongue.gif[/img] [img]tongue.gif[/img]

    Tha really stinks I think.

  10. #10
    Guest
    I dont use overture except for research. Im sure it is overrun with webmasters and site owners that are just checking things out. Of course it is inflated.

  11. #11
    Newbie
    Join Date
    January 18th, 2005
    Posts
    29
    BUT..........even if all the stats are inflated 10x, the relative popularity should still have integrity.

  12. #12
    Resident Genius and Staunch Capitalist Leader's Avatar
    Join Date
    January 18th, 2005
    Location
    Florida
    Posts
    12,817
    Unfortunately, ALL the stats there at Over are NOT inflated "equally". Rather, only the terms a lot of webmasters are interested in checking, and those that appear on partner sites, are inflated. Some terms are usually not inflated at all--which can make HUGE gaps in the apparent popularity.

    Eventually you can get a feel for which stats are legit and which ones aren't, but this requires quite a bit of trial and error, while taking note of what really happens (traffic-wise) with terms you have bid on.

    I will note that one other (legit) reason Over's results are generally higher than WordTracker's is that Over measures monthly while WT's prediction is for daily. This makes a drastic-looking difference between the two...

    But occasionally someone will sic a "rank-checking" robot on Dogpile/Metacrawler so some really hokum terms can come up as popular at WT, as well. But inflation doesn't happen nearly as often as it does at Over. Probably not too many people are interested in checking their own ranks at Dogpile!

    As for bidding more at Over due to the expectations of higher traffic, that's a recipe for doom!! I've seen more than one term where the top bidders have bid it up so high it could no longer possibly be profitable, whether they got 1 click or a million. It can be rather fun to watch competitors commit business suicide in this manner...

    FOR A FREE TOOL, Over's was still the best one the last time I looked, but WordTracker (which charges) does beat it by a wide margin in terms of accuracy.

  13. Newsletter Signup

+ Reply to Thread

Similar Threads

  1. new from Wordtracker
    By annic in forum Midnight Cafe'
    Replies: 4
    Last Post: April 14th, 2006, 11:52 AM
  2. overture keyword tool Vs Wordtracker
    By oscar in forum Search Engine Optimization
    Replies: 5
    Last Post: May 12th, 2005, 05:59 AM
  3. Replies: 8
    Last Post: April 11th, 2004, 09:58 PM
  4. Replies: 0
    Last Post: April 10th, 2004, 03:18 PM
  5. Overture stock takes a tumble
    By ecomcity in forum Building Traffic, Newsletters & Advertising
    Replies: 0
    Last Post: February 7th, 2003, 10:33 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •